In the last few years, the frequency of flood events has highlighted the need to adopt measures that can enable communities to deal with them and reduce the damage they cause. It is essential to adopt an approach that involves improving decision-making by drawing on accurate, qualified and real-time information. This is because a decision that is not carried out correctly or based on erroneous data, runs the risk of causing serious problems and irreversible damage. However, current available decision support systems for flood management are largely based on static and limited data, and as a result, these systems usually fail to handle volunteered geographic information and data streams from heterogeneous sources in real time. In addition, there is a lack of approaches which can systematically incorporate decision support systems into decision-making procedures for flood management. Because of this, it is essential to structure and define the different stakeholders involved and the specific information needed, as well as a set of appropriate features, and different visualization requirements to improve the accuracy and increase the speed of the decision-making.
This project seeks to fill in these gaps, by setting out a decision-based integration approach with the aim of combining heterogeneous data sources (the information of volunteers, together with that provided by static sensors) and supporting decision-making in real-time flood management. For this reason, it defines a set of approaches which are responsible for the following: handling the information needed to support decision-making (from emergency agencies to volunteers), evaluating the quality of the data provided by the volunteers, integrating this information with that collected by sensors, and providing technological decision support resources that can enable the decision- makers to analyze this information in real-time. Furthermore, this approach will involve evaluating the use of case studies, conducting controlled experiments, and simulating the business processes concerned with flood risk management in São Carlos/SP.
Furthermore, the results obtained on this project have been contributing for the following tool.
GeoDashboard is a web-based dashboard which aims to support real-time decision-making by a set of performance indicators that are essential to flood risk monitoring. This tool is built upon open-source frameworks and geoservices that follow the standards of Open Geospatial Consortium’s (OGC). We deployed the prototype to handle data gathered from a real Wireless Sensor Network (WSN) in the city of São Carlos.